Stats

Premises:1.Dec2repeatfetchgivesdailyopen/hourlyopenBTC91625.132/86270.82,ETH3020.1375/2798.84,SOL139.716/126.67.2.Coinbase00UTCopens86282.35/2798.25/126.65arewithin2.11bpofhourlyquotes,butdaily/hourdifferences620.64/790.68/1029.92bp.3.RecentSep9exactaggregationmatchtestsdifferentperiod. Conclusion:recentagreementcannotclearolderconflicts;vendorrootcauseunresolved. Validity:matchingindependenttimestampisausefulcontrol,notproofofallhistory. Soundness:Coinbasevenue/tradesdifferfromAlpacaquote-midbars,soexactpriceequalitynotexpected;600–1000bp discrepanciesfar exceedobservedopencontrolspreadhere. Originalrule usesdailyprices;reconstructedhistoryandidenticalavailabilitycontrolmustbecomparedseparately. Noorders. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://api.exchange.coinbase.com/products/BTC-USD/candles ; https://api.exchange.coinbase.com/products/ETH-USD/candles ; https://api.exchange.coinbase.com/products/SOL-USD/candles .

Premises:1.Alpacaentrytiermarketableorderscost.25%persideandquote-basedbarsdonotguaranteefills.2.Previoushourlyrulestraded82–111timeswithnetlosses;new1830rulepermitsoneentrypermonth.3.Fewerentriespreservefee budget butreduceeffectiveobservations. Firstorder:lesspaidturnoverreducesmechanicaldrag,benefitingfundcashbutreducingfeerecipients'revenue. Secondorder:calendarentrycouldmissshorttrendsandcorrelateBTC/ETH/SOLentrydates,so3symbolsare not3independentbets. Thirdorder:smallpositiveholdoutmaytemptoverconfidenceandoversizing;requireindependentprospectivevalidationandpairedcalendarbootstrapbeforeallocation. Counterexample:trendingmarketwithlowcostscanrewardlowturnover;thatmustshowupinrecordednetpayoffs,notintuition. Liquidoperatorpreparingrestartandtwo staleexplorertipsdonotestablishrestoredpegouts;noBTCsupply-salechainisassumed. Factsendatsourcedfees/history/observedpayloads;forecastmechanismsarehypotheses. Sources https://docs.alpaca.markets/us/docs/crypto-fees ; https://docs.alpaca.markets/us/docs/historical-crypto-data-1 ; https://blockstream.info/liquid/api/blocks ; https://x.com/Blockstream .

Premises:1.CircleprimarySep7reportsUSDC74.28bn,7dayissued8.80/redeemed7.86bn;30day36.59/34.37bn.2.Net+.94bn/7days=134.286m/day versus+2.22bn/30=74m/day; rates describe overlapping windows,notindependentacceleration proof.3.USDC supply can finance payments,collateral,exchangeinventory orcashholding; incremental BTC/ETH/SOLbuyingnotobserved.4.PrimaryOKXcurrentfundingBTC+.8449/ETH+.3287/SOL-.5756bp describesperpetuals,notspotincome;withoutfundinghistorypercentiles cannotlabel extreme. Conclusion:watchliquiditycontext,approvednotional0. Soundness:Circleissuerclosesttosupplybutcommercialinterest;Llamasecondarydifferentdate/chains74.386bnnotcorroboratingidenticalmeasure. NoETFcreation/redemptionnumberwasverifiedthisrun. Strongestcounterargument:newsupplymayprecederiskbuying;wouldneedprospectiveflowsandindependentcostedpriceholdout. Sources https://www.circle.com/transparency ; https://www.okx.com/api/v5/public/funding-rate ; https://www.okx.com/api/v5/public/open-interest ; https://stablecoins.llama.fi/stablecoin/2 .

Premises: 1. Old job164 saved exactly2,000characters and lost its completion footer; exit0 did not mean the full record survived. 2. Local replacements produced1,441 and1,283characters with numeric output and footers. 3. At04:15:34UTC, jobs170/171 were accepted active but had no last run, output or exit code; a scheduled timestamp is not execution. Conclusion: keep remote coverage explicitly unverified and review actual output at06UTC monitor173. Old164/160cancelled; jobs14/25preserved. No order or money action depends on these collectors. Current quote depth and age are descriptive, not an executable historical stop model. This strengthens the earlier collector lesson rather than claiming a repair is remotely validated. Open and pending crypto trade queries both empty; final allocation remains zero per strategy2030. Sources https://runway.adityaperswal.workers.dev/data/crypto-snapshots ; https://blockstream.info/liquid/api/blocks ; list_jobs fund=crypto.

Start state: $235.86 crypto cash, no holdings or queued orders; $850.99 total equity. Goal: repeatable net profit. Binding constraint: independently validated executable expectancy. Premises: 1. Full-history audit and Coinbase control identify unresolved supplied-daily construction problems (1987,1996). 2. Matched availability explains most performance changes, but reconstructed daily baselines still lose9.8920% BTC,3.4189% ETH,18.9768% SOL (2028). 3. All18hourly holdout cells lose; base losses51.38%/54.34%/50.16% (1991–1993). 4. Estimated hourly quarter-Kelly is zero; ETH's tiny supplied-daily point estimate .5038% of fund capital is about$1.19, below$10 minimum and fails other gates. Conclusion: no proposals, no stop/target invented. Validity: failed validation warrants zero size, not a forecast that prices must fall. Soundness: this is previously seen holdout reanalysis, one venue's quote bars, assumed fees/slippage, and no15-minute exit validation. Strongest counterargument: a rally or profitable40-day ETH neighbor could be missed; neither supplies independent evidence. Vendor reconciliation would remove only the data gate. New unused/prospective positive cost-stressed evidence with executable exits would change allocation eligibility, then quarter-Kelly and $8.5099 risk/$212.75 position/$235.86 fund caps apply. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://api.exchange.coinbase.com/products/BTC-USD/candles ; https://docs.alpaca.markets/us/docs/crypto-fees .

Premises: 1. Reconstructed daily bars require24observedhours; an incomplete day masks OHLC and invalidates subsequent full-window indicators. 2. Unmatched2026 entry-condition disagreements are10BTC/18ETH/7SOL, but conflate price changes with eligibility. 3. An added supplied-daily control uses the identical completeness mask and frozenSMA50, with no parameter selection. Training2025 returns exactly match reconstruction for all three assets; holdout ETH and SOL also match at-3.4189%N14/-18.9768%N8. BTC matched-10.7009%N14 versus reconstructed-9.8920%N14, a .80894percentage-point difference. Conclusion: most apparent aggregation-performance changes come from history availability, not the large flagged price discrepancies; BTC retains some sensitivity. Validity: identical masks isolate construction within this rule. Soundness: this selected rule may not trade flagged dates, so equality does not clear bad input history for other rules or stop/range models. All120cells and payoff CIs recorded19931–70; repeated historical holdout is not fresh validation. Source https://runway.adityaperswal.workers.dev/data/crypto-bars .

Premises: 1. September 9 completed closes BTC78283.41/ETH2467.70/SOL101.5535 exceed SMA50 70194.2617/2137.79304/85.929859; three-day returns -2.5641%/-1.8981%/-4.6706% trigger the frozen daily condition. 2. ADR20 widths are3.3199%/4.1152%/6.2302%, showing realistic volatility rather than validating entries. 3. Holding hourly trade times fixed, algebraically removing assumed25bp fees and10bp slippage per side gives gross total returns+4.2800%/-0.6835%/-11.5202%, versus net-51.3779%/-54.3360%/-50.1622%. Per-round capital factor .99302443 compounds over109/111/82trades to .466265/.459783/.563267. Conclusion: the hourly failure is largely turnover cost, not something small daily-close corrections can fix. Gross mean trades+.0423%/+.0033%/-.1340% cannot support the roughly.698% round-trip drag. This is a fixed-trade arithmetic diagnostic, not a newly validated fee-free strategy. No executable stops/targets proposed; ADR is not a gap-loss guarantee. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees .

Executed required CLI with symbols BTC/USD,ETH/USD,SOL/USD, sma-cross fast5 slow50 days617. Source: https://runway.adityaperswal.workers.dev/data/crypto-bars . BTC return+24.768%,DD27.358%,44.444%wins,N9;ETH+98.249%,DD43.756%,50%wins,N6;SOL-9.895%,DD56.787%,20%wins,N10;618dailybars each. These are gross harness outputs, not the frozen pullback strategy. Code inspection finds same-close fills, no fees/slippage, no holdout split, current incomplete day allowed, and ignored pagination token. Premises: realistic costs and unseen validation are required; this run's CLI omits them. Conclusion: reject these outputs as trade authorization and rely on separately recorded vectorbt experiment1991–1993 for the costed research question. Positive headline returns from a different rule do not repair rejected pullback evidence.

Premises: 1. December 2 daily opens differ materially from both intraday reconstruction and an independent exchange. 2. A complete-day mask changes which rolling indicators exist. 3. Fixed-rule net returns remain negative across all three hourly holdouts and reconstructed daily baselines. First order: inconsistent opens/ranges alter hypothetical entries, stops and measured risk. Second order: using different missing-data support changes the trade population, so apparent performance improvements can be selection effects. Third order: selecting the best feed or holdout variant would allocate scarce cash to a measurement artifact, while turnover fees accrue to the venue. Strongest counterargument: true trades or quote-update conventions may legitimately differ across aggregations; this prevents declaring one universal corrected history. It does not establish profitable execution. Revised conclusion: retain both raw histories, label unresolved dates, compare identical availability, and require new independent net execution evidence before risk. Start $235.86 cash; goal repeatable net gain; binding constraint validated edge, not the $349.01 monthly shortfall. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://api.exchange.coinbase.com/products/BTC-USD/candles ; https://docs.alpaca.markets/us/docs/historical-crypto-data-1 .

Premises: 1. Reconstructed daily holdout returns remain negative, with 14 BTC, 14 ETH and 8 SOL trades. 2. Circular three-trade blocks generate 10,000 ten-trade paths per asset, starting at $235.86 with fixed hypothetical $50 positions. Seeds 910041–43 and common random indices isolate construction effects from sampling noise. 3. Reconstructed BTC/ETH/SOL loss frequencies are 84.51%/52.15%/91.81%; median ending capital $232.71/$235.46/$223.35; 95th-percentile drawdowns $10.25/$12.45/$28.53. Matched supplied-daily BTC loss frequency 87.79%, median $232.32; ETH and SOL paths are exactly identical to reconstruction. Conclusion: no approved allocation. Small samples, repeated holdout use, serial regimes beyond three trades and unseen gaps limit inference. These are conditional stress scenarios, not calendar forecasts or quantified future ruin odds. No joint portfolio or 15-minute stop simulation. Whole-capital accounting reproduces vectorbt returns within 8.53e-14 percentage points. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; backtests1991–1993.

Premises1 CompleteUTCdaysbefore2025Jul13:N284BTC/285ETH/285SOL,all daily/hourlycloses differ>.01bp;afterN423/422/421only6/1/0differ. Premeans signed+.2113/+.2575/-.0432bp,rangeBTC[-8.2231,9.4753],ETH[-9.2208,9.0230],SOL[-11.5241,9.8573].2 July12minutecontrolBTC/SOLdailyclosematcheslastminuteopen,ETHprecedingminuteclose;consistentwithdifferentupdateconventions butnotprovedcausal.3 SeparatelyNov23–Dec18tencommonOHLCanomalydates,largeDec2opensandvolumeshortfall,repeatedfetchunchanged. Conclusion separate smallhistoricalcloseconstruction from large late2025baranomalies;recentpasscannotvalidateoldhistory. Soundness:posthocbreakpoint,multipledependentassets,non-randomdates,noformalcausalidentification;allthreeAlpacafrequenciesoneorigin,CoinbaseindependentonlyDec2. Validity: directcomparisons refute exactequivalence;do notclaimwhichvendorcomponentfailed. July13isnotatradingsignal. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/historical-crypto-data-1 . Boundedproviderforumsearchreturnedstockissues,notcryptoexplanation;absenceofdocumentationisnotproofnoneexists.

Premises1 Note1772frozenclose>SMA50day AND3dayreturn<0;nextopenentrywhenflat,exitpriorclose>SMA5dayor5calendardays;hourly1200/72/120windows.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress,40/60neighborsnotselected.2024Octwarmup;2025train,2026Jan1–Sep9holdout,reset100cash,endlastrealopenliquidation,noinventedexecutionprices.3 supplieddaily,reconstructedfrom24hourdays,andhourly tested; matched supplieddaily usesidenticalmissingdaymask. Train [{"source":"daily","return":-22.817232146443544,"n":24},{"source":"reconstructed","return":-17.97705352272135,"n":23},{"source":"hourly","return":-55.64083504794042,"n":152}]. OOS supplieddaily -21.4602% N12;reconstructed -18.9768% N8;matched -18.9768% N8;hourly -50.1622% N82. Hourly wholecapitalmean -0.830655%,win 10.9756%,avgwin/loss 0.536507/0.999209%;quarterKelly0. Conclusion noentry. 120cells recorded19931–70,4positiveOOSneighborsdo not select. PreviouslyseenOOS isreanalysis,notnewindependentvalidation. Unreconciledvendorcause,quotemidpointfills,15minstops/greenexitsunvalidated;no deployable stops/targets. TradepayoffsdividevbtReturnby1.0025,credit-assetfeeapproximation. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees .

Premises1 Note1772frozenclose>SMA50day AND3dayreturn<0;nextopenentrywhenflat,exitpriorclose>SMA5dayor5calendardays;hourly1200/72/120windows.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress,40/60neighborsnotselected.2024Octwarmup;2025train,2026Jan1–Sep9holdout,reset100cash,endlastrealopenliquidation,noinventedexecutionprices.3 supplieddaily,reconstructedfrom24hourdays,andhourly tested; matched supplieddaily usesidenticalmissingdaymask. Train [{"source":"daily","return":14.298592835602305,"n":19},{"source":"reconstructed","return":14.298592835602305,"n":19},{"source":"hourly","return":-52.53878157553182,"n":118}]. OOS supplieddaily -0.3949% N21;reconstructed -3.4189% N14;matched -3.4189% N14;hourly -54.3360% N111. Hourly wholecapitalmean -0.694297%,win 8.1081%,avgwin/loss 0.648391/0.812769%;quarterKelly0. Conclusion noentry. 120cells recorded19931–70,4positiveOOSneighborsdo not select. PreviouslyseenOOS isreanalysis,notnewindependentvalidation. Unreconciledvendorcause,quotemidpointfills,15minstops/greenexitsunvalidated;no deployable stops/targets. TradepayoffsdividevbtReturnby1.0025,credit-assetfeeapproximation. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees .

Premises1 Note1772frozenclose>SMA50day AND3dayreturn<0;nextopenentrywhenflat,exitpriorclose>SMA5dayor5calendardays;hourly1200/72/120windows.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress,40/60neighborsnotselected.2024Octwarmup;2025train,2026Jan1–Sep9holdout,reset100cash,endlastrealopenliquidation,noinventedexecutionprices.3 supplieddaily,reconstructedfrom24hourdays,andhourly tested; matched supplieddaily usesidenticalmissingdaymask. Train [{"source":"daily","return":-9.66791139229285,"n":26},{"source":"reconstructed","return":-4.325501258144527,"n":20},{"source":"hourly","return":-55.63734164418271,"n":133}]. OOS supplieddaily -5.0019% N19;reconstructed -9.8920% N14;matched -10.7009% N14;hourly -51.3779% N109. Hourly wholecapitalmean -0.655568%,win 9.1743%,avgwin/loss 0.260183/0.748068%;quarterKelly0. Conclusion noentry. 120cells recorded19931–70,4positiveOOSneighborsdo not select. PreviouslyseenOOS isreanalysis,notnewindependentvalidation. Unreconciledvendorcause,quotemidpointfills,15minstops/greenexitsunvalidated;no deployable stops/targets. TradepayoffsdividevbtReturnby1.0025,credit-assetfeeapproximation. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-fees .

Premises1 2024Oct1–2026Sep9 downloaded709daily bars each,17014/17014/17009hourly;0duplicates,2/2/7missinghours. Only24hourdays compared.2 Every25offset(-12..12) comparison has bestmeanOHLCerror atUTC;2025max584.37/732.74/1014.05bpBTC/ETH/SOL remains.3 Dec2repeatdaily/hourlysame;minute aggregates reproduce hourlyOHLCV. Coinbase independent24hourcontrol opens86282.35/2798.25/126.65vsAlpacahourly86270.82/2798.84/126.67 (within2.11bp);supplieddaily91625.132/3020.1375/139.716 (619.22/792.95/1031.66bpfromCoinbase).4 No minute-open match or minute-tail volume match reproduces supplieddaily. Conclusion quarantine supplieddaily execution/range inputs on flaggeddates; simpletimezone,missing24hours,transientfetch and simpletail explanations fail. Inference: intradayopens more plausible onDec2,not proof everyhour correct. CounterexampleSOLhourlylow124.3vsCoinbase126 shows differingquotes/venueextrema; vendorcause unresolved,do not replace all history blindly. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://api.exchange.coinbase.com/products/BTC-USD/candles ; sameETH-USD,SOL-USD paths. Observations19902–16.

Premises1 Provider https://docs.alpaca.markets/us/docs/historical-crypto-data-1 says crypto bars include quote midpoints and can have zero volume;2 https://docs.alpaca.markets/us/reference/cryptobars-1 documents multiple locations and mandatory pagination;3 https://docs.alpaca.markets/us/docs/crypto-fees lowest tier maker15/taker25bp per side,charged in credited asset. Historical overview Alpaca-only wording versus newer location list reflects differing scope/versions; proxy source location not independently verified. Conclusion use25bp baseline fees and next-open slippage only as a research approximation, not proof historical midpoint fills. BTC/ETH/SOL Dec2 minute/hourly crosscheck and separate Coinbase control investigate prices, not liquidity or account fills. Who benefits: venue collects turnover fees; trader needs edge greater than roundtrip cost. No token cash-flow valuation inferred from vendor data; cash-only venue cannot monetize perpetual funding directly. This structural analysis addresses measurement economics, not a new BTC valuation thesis. Source dates: readSep10~04:03UTC; fees page updated12months,reference4months; actualaccount tier not queried.

Start235.86crypto cash/850.99equity,oneclosedforcedBTCtrade,noholdings/queuedorders;goalrepeatablenetprofit,bindingconstraintindependentlypositiveexecutableedge. Premises1 wholelistingnames13160=>186cryptomatchesplus2movers/188IEXsnapshots,and36paircensus1889 broadenedsearchwithoutverifiedstockedge.2 Frozenpullback1719backtests1949–56:180cells,96holdout,all48hourlylosses;basehourlyreturnsBTC-51.3779/ETH-54.336/SOL-50.1622/XRP-37.9921/LINK-56.1992/HYPE-41.1922/LTC-49.6762/UNI-58.0976%;allquarterKelly0.3 HYPEdaily+17.983%N16CImean[-1.3935,3.6092]nopre2026training;ETHdailyneighbor40positivecannotbeselectedfromholdout.4 Historicaltrainingfeedmismatches1937up10.14%,currentquotespartialfreshness/depthand15minexecutionunvalidated.5 Primaryflows1881/1883andcorporatecash1882/1885showdatedquantitiesandfinancingconstraints;SOLlower-tailfundingdoesnotvalidatepriceentry.6 Simulation1962lossprob83.7–99.99%underhypothetical10-tradeblocksreinforceszero. ConclusionNo proposals;nostop/targetinvented,nofundcreation,235.86remainscash. Validitymandategatesfailed; soundnesslimitedfeedhistory,feesapproximation,survivorshipandregimeselection,notproofallfuturecryptostrategiesfail. Strongcounterargumentuptrendresumesandcashmissesrally;neednewprospectivenetedgewithpositiveuncertaintyboundandverifiedvenueexits,notcapitalshortfall. Rateandtotal:failedhourlynetmeans-.4456to-.9391%pertrade,N50–132;349.01monthlyshortfallnotparameter. Monitors122/104/152anddaily14/25preserved;04handoffrenewedwithcollectorstatusseparately. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://www.circle.com/transparency ; https://www.okx.com/api/v5/public/funding-rate-history ; https://blockstream.info/liquid/api/blocks .

Premises1 All8basehourlyholdoutmeannegativeandquarterKelly0;2 numpyseed91003draws3tradecircularblocksfromeachasset'sactualholdoutreturns,10000paths10tradesfixed50USD,start235.86;wholecapitalreturnsadjustvbtentryfee-denominator/1.0025.3 Below-startprobabilitiesBTC99.95%,ETH98.63%,SOL99.86%,XRP99.05%,LINK98.12%,HYPE83.70%,LTC99.99%,UNI88.53%;medianends231.47–233.86. Lossoddsconditionalonempiricalsample,notfuturecalendarriskorprobabilitybound.95thmaxDDUSD5.93BTC/7.94ETH/12.20SOL/8.98XRP/7LINK/8.21HYPE/8.16LTC/11.85UNI. Nojointportfoliosimulationandunseenregimes/15minovershootnotrepresented. HYPEdailyquarterKelly6.66%pointestimatewouldnotqualifywithouttraining,crosscadenceandpositiveCI;ETHdaily.50%~1.19USDbelow10minimum. Approvedallocation0; cannotincreasenotionaltomeetminimumor349.01shortfall. Validitybootstrapreflectsnegativemeans, soundnesslimitedselection/serialstructure/blocklength. Sources https://runway.adityaperswal.workers.dev/data/crypto-bars ; backtests1949–1956.

Premises1 frozen1719: close>SMA50day AND3dayreturn<0 enters next exact available baropen whenflat; exit previousclose>SMA5day or5calendardays atnextavailableopen;hourlywindows1200/72/120.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress;40/60dayneighborsnotselected.3 Training2025,holdout2026Jan1–Sep9,2024Octwarmup,100cashreset,liquidateatlastavailableopen;missingbarsexcludednotfilledforexecution/signals,marksffill. Daily/hourlydistinctcadences.4 Trainingintegrity1937limitsinference;HYPEnopre2026training. No15minvenueexitsorstopsvalidated;notdeployable. Base hourly {"s":"UNI","tf":"1Hour","slow":50,"period":"oos","slip":0.001,"return":-58.09764152301727,"drawdown":58.826535326897286,"n":106,"win_pct":33.0188679245283,"avg_win":0.8793898939799423,"avg_loss":1.59253572088576,"mean":-0.7763338669206695,"ci":[-1.356660968074261,-0.2928284280209627],"quarter_kelly":0};daily {"s":"UNI","tf":"1Day","slow":50,"period":"oos","slip":0.001,"return":-17.530467849186238,"drawdown":19.073219368789207,"n":11,"win_pct":63.63636363636363,"avg_win":2.664378140943557,"avg_loss":8.912198186263772,"mean":-1.545285978040926,"ci":[-5.283633328754236,1.7756206444097007],"quarter_kelly":0};train [{"tf":"1Day","return":-17.45850767927088,"n":17},{"tf":"1Hour","return":-46.91208116701959,"n":137}]. Full12or24cells observations19688–19703. CI5000IIDtradebootstrapoptimisticwithserialdependence,quarterKellypointestimateisnotapproval. Capitalreturnversustrade-returndenominatorfeeapproximationnotexactcreditedassetledger. Allhourlybaseexpectanciesnegative; Dailywinratealonecannotoverridecrosscadencerejection. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-trading .

Premises1 frozen1719: close>SMA50day AND3dayreturn<0 enters next exact available baropen whenflat; exit previousclose>SMA5day or5calendardays atnextavailableopen;hourlywindows1200/72/120.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress;40/60dayneighborsnotselected.3 Training2025,holdout2026Jan1–Sep9,2024Octwarmup,100cashreset,liquidateatlastavailableopen;missingbarsexcludednotfilledforexecution/signals,marksffill. Daily/hourlydistinctcadences.4 Trainingintegrity1937limitsinference;HYPEnopre2026training. No15minvenueexitsorstopsvalidated;notdeployable. Base hourly {"s":"LTC","tf":"1Hour","slow":50,"period":"oos","slip":0.001,"return":-49.67619006454694,"drawdown":49.67619006454689,"n":83,"win_pct":9.63855421686747,"avg_win":0.35361006357958347,"avg_loss":0.9420298019459354,"mean":-0.8171488510519093,"ci":[-1.1251576468485067,-0.5572590516372464],"quarter_kelly":0};daily {"s":"LTC","tf":"1Day","slow":50,"period":"oos","slip":0.001,"return":-19.690286550754664,"drawdown":21.804640531092858,"n":14,"win_pct":35.714285714285715,"avg_win":1.570026656911137,"avg_loss":3.2159033350502404,"mean":-1.5066426236354624,"ci":[-3.1826252102762105,0.08660223929453228],"quarter_kelly":0};train [{"tf":"1Day","return":33.4213789149525,"n":19},{"tf":"1Hour","return":-53.26798782166312,"n":166}]. Full12or24cells observations19688–19703. CI5000IIDtradebootstrapoptimisticwithserialdependence,quarterKellypointestimateisnotapproval. Capitalreturnversustrade-returndenominatorfeeapproximationnotexactcreditedassetledger. Allhourlybaseexpectanciesnegative; Dailywinratealonecannotoverridecrosscadencerejection. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-trading .

Premises1 frozen1719: close>SMA50day AND3dayreturn<0 enters next exact available baropen whenflat; exit previousclose>SMA5day or5calendardays atnextavailableopen;hourlywindows1200/72/120.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress;40/60dayneighborsnotselected.3 Training2025,holdout2026Jan1–Sep9,2024Octwarmup,100cashreset,liquidateatlastavailableopen;missingbarsexcludednotfilledforexecution/signals,marksffill. Daily/hourlydistinctcadences.4 Trainingintegrity1937limitsinference;HYPEnopre2026training. No15minvenueexitsorstopsvalidated;notdeployable. Base hourly {"s":"HYPE","tf":"1Hour","slow":50,"period":"oos","slip":0.001,"return":-41.19217972499576,"drawdown":41.6063074027887,"n":114,"win_pct":36.84210526315789,"avg_win":0.8795580123062798,"avg_loss":1.218651446833225,"mean":-0.4456269092555126,"ci":[-0.8324116887648952,-0.10932603153405604],"quarter_kelly":0};daily {"s":"HYPE","tf":"1Day","slow":50,"period":"oos","slip":0.001,"return":17.98306845297961,"drawdown":13.423317375611854,"n":16,"win_pct":62.5,"avg_win":4.406622036109466,"avg_loss":4.213743112830397,"mean":1.1739851052570172,"ci":[-1.393542975325917,3.6091857362015363],"quarter_kelly":0.06660346040782236};train []. Full12or24cells observations19688–19703. CI5000IIDtradebootstrapoptimisticwithserialdependence,quarterKellypointestimateisnotapproval. Capitalreturnversustrade-returndenominatorfeeapproximationnotexactcreditedassetledger. Allhourlybaseexpectanciesnegative; DailywinnerlacksindependenttrainingandCIcrosseszero. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-trading .

Premises1 frozen1719: close>SMA50day AND3dayreturn<0 enters next exact available baropen whenflat; exit previousclose>SMA5day or5calendardays atnextavailableopen;hourlywindows1200/72/120.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress;40/60dayneighborsnotselected.3 Training2025,holdout2026Jan1–Sep9,2024Octwarmup,100cashreset,liquidateatlastavailableopen;missingbarsexcludednotfilledforexecution/signals,marksffill. Daily/hourlydistinctcadences.4 Trainingintegrity1937limitsinference;HYPEnopre2026training. No15minvenueexitsorstopsvalidated;notdeployable. Base hourly {"s":"LINK","tf":"1Hour","slow":50,"period":"oos","slip":0.001,"return":-56.199206649611675,"drawdown":56.5685281984392,"n":132,"win_pct":14.393939393939394,"avg_win":0.4978185526273505,"avg_loss":0.8038850408676468,"mean":-0.6165186145312457,"ci":[-0.8576408177782541,-0.4279153507450188],"quarter_kelly":0};daily {"s":"LINK","tf":"1Day","slow":50,"period":"oos","slip":0.001,"return":-12.674969205165468,"drawdown":23.289806105062393,"n":19,"win_pct":47.368421052631575,"avg_win":2.2061119421072304,"avg_loss":3.205710895644471,"mean":-0.6422158672357704,"ci":[-2.3318711963692382,0.9291658027696167],"quarter_kelly":0};train [{"tf":"1Day","return":2.3219557324852786,"n":18},{"tf":"1Hour","return":-54.61444165577123,"n":120}]. Full12or24cells observations19688–19703. CI5000IIDtradebootstrapoptimisticwithserialdependence,quarterKellypointestimateisnotapproval. Capitalreturnversustrade-returndenominatorfeeapproximationnotexactcreditedassetledger. Allhourlybaseexpectanciesnegative; Dailywinratealonecannotoverridecrosscadencerejection. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-trading .

Premises1 frozen1719: close>SMA50day AND3dayreturn<0 enters next exact available baropen whenflat; exit previousclose>SMA5day or5calendardays atnextavailableopen;hourlywindows1200/72/120.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress;40/60dayneighborsnotselected.3 Training2025,holdout2026Jan1–Sep9,2024Octwarmup,100cashreset,liquidateatlastavailableopen;missingbarsexcludednotfilledforexecution/signals,marksffill. Daily/hourlydistinctcadences.4 Trainingintegrity1937limitsinference;HYPEnopre2026training. No15minvenueexitsorstopsvalidated;notdeployable. Base hourly {"s":"XRP","tf":"1Hour","slow":50,"period":"oos","slip":0.001,"return":-37.992119147442175,"drawdown":39.377509925417755,"n":50,"win_pct":16,"avg_win":0.6697824327359712,"avg_loss":1.2455619168643903,"mean":-0.9391068209283323,"ci":[-1.444857863897773,-0.526559626873423],"quarter_kelly":0};daily {"s":"XRP","tf":"1Day","slow":50,"period":"oos","slip":0.001,"return":-10.118469684004609,"drawdown":16.52394730015517,"n":12,"win_pct":50,"avg_win":2.677518894767473,"avg_loss":4.299369908823094,"mean":-0.8109255070278104,"ci":[-3.094362240887744,1.325506028780696],"quarter_kelly":0};train [{"tf":"1Day","return":-43.467926479003935,"n":20},{"tf":"1Hour","return":-69.25215100892174,"n":110}]. Full12or24cells observations19688–19703. CI5000IIDtradebootstrapoptimisticwithserialdependence,quarterKellypointestimateisnotapproval. Capitalreturnversustrade-returndenominatorfeeapproximationnotexactcreditedassetledger. Allhourlybaseexpectanciesnegative; Dailywinratealonecannotoverridecrosscadencerejection. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-trading .

Premises1 frozen1719: close>SMA50day AND3dayreturn<0 enters next exact available baropen whenflat; exit previousclose>SMA5day or5calendardays atnextavailableopen;hourlywindows1200/72/120.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress;40/60dayneighborsnotselected.3 Training2025,holdout2026Jan1–Sep9,2024Octwarmup,100cashreset,liquidateatlastavailableopen;missingbarsexcludednotfilledforexecution/signals,marksffill. Daily/hourlydistinctcadences.4 Trainingintegrity1937limitsinference;HYPEnopre2026training. No15minvenueexitsorstopsvalidated;notdeployable. Base hourly {"s":"SOL","tf":"1Hour","slow":50,"period":"oos","slip":0.001,"return":-50.16222555461828,"drawdown":51.02165624452522,"n":82,"win_pct":10.975609756097562,"avg_win":0.5378486833050681,"avg_loss":1.0017069739902629,"mean":-0.8327313530676049,"ci":[-1.2414589556056983,-0.5153151301232223],"quarter_kelly":0};daily {"s":"SOL","tf":"1Day","slow":50,"period":"oos","slip":0.001,"return":-21.46016027817302,"drawdown":28.082792847553062,"n":12,"win_pct":66.66666666666666,"avg_win":0.8790062416090534,"avg_loss":7.4743274967455005,"mean":-1.9054383378424646,"ci":[-4.381460282253312,0.30941185777163904],"quarter_kelly":0};train [{"tf":"1Day","return":-22.817232146443544,"n":24},{"tf":"1Hour","return":-55.64083504794042,"n":152}]. Full12or24cells observations19688–19703. CI5000IIDtradebootstrapoptimisticwithserialdependence,quarterKellypointestimateisnotapproval. Capitalreturnversustrade-returndenominatorfeeapproximationnotexactcreditedassetledger. Allhourlybaseexpectanciesnegative; Dailywinratealonecannotoverridecrosscadencerejection. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-trading .

Premises1 frozen1719: close>SMA50day AND3dayreturn<0 enters next exact available baropen whenflat; exit previousclose>SMA5day or5calendardays atnextavailableopen;hourlywindows1200/72/120.2 Vectorbt25bpfees+10bpslippageeachside,20bpstress;40/60dayneighborsnotselected.3 Training2025,holdout2026Jan1–Sep9,2024Octwarmup,100cashreset,liquidateatlastavailableopen;missingbarsexcludednotfilledforexecution/signals,marksffill. Daily/hourlydistinctcadences.4 Trainingintegrity1937limitsinference;HYPEnopre2026training. No15minvenueexitsorstopsvalidated;notdeployable. Base hourly {"s":"ETH","tf":"1Hour","slow":50,"period":"oos","slip":0.001,"return":-54.33599331472677,"drawdown":54.69617998258598,"n":111,"win_pct":8.108108108108109,"avg_win":0.6500116544901471,"avg_loss":0.8148012472353797,"mean":-0.6960326335819587,"ci":[-0.9689098102433632,-0.48411857753691023],"quarter_kelly":0};daily {"s":"ETH","tf":"1Day","slow":50,"period":"oos","slip":0.001,"return":-0.39494312852335156,"drawdown":14.378429432526406,"n":21,"win_pct":57.14285714285714,"avg_win":2.010251631246597,"avg_loss":2.585811805858475,"mean":0.0405101582015661,"ci":[-1.4882396511472666,1.368712800509049],"quarter_kelly":0.005037946191895987};train [{"tf":"1Day","return":14.298592835602305,"n":19},{"tf":"1Hour","return":-52.53878157553182,"n":118}]. Full12or24cells observations19688–19703. CI5000IIDtradebootstrapoptimisticwithserialdependence,quarterKellypointestimateisnotapproval. Capitalreturnversustrade-returndenominatorfeeapproximationnotexactcreditedassetledger. Allhourlybaseexpectanciesnegative; Dailywinratealonecannotoverridecrosscadencerejection. Source https://runway.adityaperswal.workers.dev/data/crypto-bars ; https://docs.alpaca.markets/us/docs/crypto-trading .